Get an output snapshot of an agent job
Agent Endpoints
Get Agent Snapshot
GET
Get an output snapshot of an agent job
While an agent runs, it version-controls its output artifacts — the structured JSON result as it takes shape, plus any markdown, HTML, text, or screenshot artifacts it produces. Every change emits an
artifact.updated trace event referencing the new version by snapshotId. This endpoint fetches the full content of one such snapshot.
What it’s for
- Partial results — preview the agent’s structured output before the run finishes. The JSON result artifact (
artifactId: "result") is snapshotted as it grows, so eachsnapshotIdis a point-in-time view of the finaldata. - Rendered artifacts — fetch the content of non-JSON artifacts the agent produced, such as a markdown report or a screenshot of a page it visited.
- Run inspection — pair with the trace to reconstruct exactly how the output evolved over the run.
How it works
Theartifact.updated event’s change field tells you how the snapshot relates to the previous one: init (first version), partial, append, modify, or update. Each change gets a new snapshotId — fetch the latest one for the current state, or walk the history to diff versions.
The snapshot field in the response is the artifact content as a string. For json artifacts it is JSON-encoded — parse it before use.
Snapshots are recorded on Spark 2 runs — which is every new run. Jobs started on Spark 1 models before their retirement have no snapshots and return
400.Are you an AI agent that needs a Firecrawl API key? See firecrawl.dev/agent-onboarding/SKILL.md for automated onboarding instructions.
Authorizations
Bearer authentication header of the form Bearer <token>, where <token> is your auth token.
Path Parameters
The ID of the agent job
The ID of the snapshot, from an artifact.updated trace event

